Data-driven and Almost Model-independent Reconstruction of Modified Gravity

TL;DR

This paper reconstructs a modified gravity parameter using Gaussian processes from current cosmic data, revealing the need for more complex models beyond standard general relativity at low redshifts.

Contribution

It introduces a data-driven, nearly model-independent method to reconstruct the modified gravity factor using Gaussian processes and current cosmic observations.

Findings

Reconstructed a time-varying modified gravity parameter at low redshifts.

Indicated that simple models like general relativity are insufficient to explain the data.

Suggested the necessity for more complex modified gravity theories.

Abstract

In this paper, a modified factor $\mu$, which characterizes modified gravity in the linear matter density perturbation theory, is reconstructed in a data-driven and almost model-independent way via Gaussian process by using currently available cosmic observations. Utilizing the Pantheon+ SNe Ia samples, the observed Hubble parameter $H(z)$ and the redshift space distortion $f\sigma_8(z)$ data points, one finds out a time varying $\mu$ at low redshifts. The reconstructed $\mu$ implies that more complicated modified gravity beyond the simplest general relativity and the Dvali-Gabadadze-Porrati braneworld model is required.